Skip to main content

Full text of "Molecular evidence for a Jurassic origin of ants."

See other formats



Naturwissenschaften 84. 22-23 (1997) © Springer- Verlag 1997 

Molecular Evidence for a Jurassic Origin of Ants 

R.H. Crozier. L.S. Jermiin*. M. Chiotis 

School of Genetics and Human Variation. La Trobe University. Bundoora. 
Victoria 3083. Australia 

* John Curtin School of Medical Research. Australian National University, 
Canberra, ACT 2601, Australia 

The ants, a group of ecologically im- 
portant insects [1], are commonly 
found in Oligocene and Miocene am- 
bers, whereas fossils of apparent ant- 
wasp intermediates are well known 
from the mid-Cretaceous [2]. Based 
on this record, it has become com- 
monly accepted that the ants arose no 
earlier than the late Cretaceous and 
that the adaptive radiation of extant 
ant genera took pJace no later than 
the start of the Tertiary [2 J, 

The recent discovery of Cariridris 
hipetiokitu. a Brazilian fossil from the 
Aptian of the early Cretaceous which 
has been placed in the ant subfamily 
Myrmeciinae [3] (represented today 
by the Mynnecia bulldog ants of Aus- 
tralia and nearby islands, has shown, 
however, that the ants diversified 
much earlier than previously thought, 
with Cariridris. an undoubted an t, 
roughly contemporaneous with the 
previously known ant-wasp intermedi- 
ates. In this review, we estimate the 
time of origin of the ant family using 
the divergences between mitochon- 
drial DNA sequences from ants of six 
subfamilies and a vespid wasp. 
We used mitochondrial cytochrome b 
sequences from Mynnecia giilosa and 
the karyotypically diverse [4] M. pilo- 
sula group [5] to calibrate the evolu- 
tionary rate of this sequence in ants. 
The evolutionary rate was calibrated 
using codon positions 1 and 2 only 
(base composition differences at posi- 
tion 3 indicate a lack of stationarity 
[6]) under the assumption that nucleo- 
tide substitutions follow Kimura's 

two-parameter model [7]. A conserva- 
tive estimate of 124.5 Ma for the age 
of Cariridris was used, this being the 
lower boundary of the Aptian, i.e., 
112-124.5 Ma ago [8]. To check the 
suitability of the data for estimating 
divergence times, we applied the rela- 
tive rate test of Wu and Li [9] (as im- 
plemented by Muse and Weir [10]), 
obtaining the results in Table 1, find- 
ing that the data do not violate the as- 
sumptions of a molecular clock. 
Using (1) this age, (2) the evolution- 
ary distances between Mynnecia spe- 
cies, and (3) the hierarchical phyloge- 
netic approach, i.e.. a method for in- 
creasing the precision in estimating 
the distance between a group of taxa 
(the M. pilosula group) and their out- 
group (M. gulosa) by taking into ac- 
count both phylogenetic structure and 
variation in component distances [11], 
yields an estimate of the evolutionary 
rate for ants of 0.165±0.023% (95% 
confidence intervals) substitutions per 
million years. 

To estimate the date of origin of the 
ants, we obtained mitochondrial DNA 
sequences homologous to those men- 
tioned above (positions 11427-12134 
in the honeybee mitochondrial gen- 
ome [12]) from five further ant sub- 
families believed to represent the 
most divergent lineages, and took the 
largest divergence value as giving the 
best estimated of the age of the 
group. This approach yields an age of 
185±36Ma ago (95% confidence 
limits derived from the SE of the di- 
vergence estimate: Fig. 1). Similar re- 

sults follow from an analysis using 
nonsynonymous substitutions [13]. 
The larger distances to the wasp, with 
a relatively narrow spread compared 
to those between ants, indicate that 
saturation has not been a serious 
problem in this calculation. 
Our finding thus places the origin of 
the ants in the early Jurassic, at least 
70 Ma earlier than recorded by fossils 
[3], and placing the group within the 


0.25 -J 
0.20 \ 
O.IS 1 
0.10 i 

X 0.05 

Million years since divergence 

Fig. 1. Sequence divergences and inferred di- 
vergence times between Mynnecia groups 
(solid circle), among the ant subfamilies 
(open circles ). and between the wasp Vespiila 
vulgaris and the ant subfamilies (solid 
squares ). Divergence times were based on an 
evolutionary rate which was determined 
using the evolutionary distances in cyto- 
chrome b sequence codon positions 1 and 2 
between species groups of living Mynnecia 
bulldog ants, and the age of the fossil bull- 
dog ant Cariridris. The most ancient diver- 
gence between ant subfamilies (here between 
Leptomyrmex an d Tetraponera) is taken as 
ifeage"or7!Te~»rotiprW!TR~9?% confidence 
intervals shown (bar). The greater spread of 
distances between ant subfamilies compared 
to that between them and the wasp reflects 
variation in divergence times between ant 
subfamilies. The Mynnecia \5] and Tetrapo- 
nera 1 18] sequences have been published pre- 
viously; new sequences, with collection 
codes and Genbank accession numbers, are 
from Camponotu s sp.. nigroaeneus gp. 
(ABZI.13. U75351), L epromyrme.x unicolor 
(ACBW.Ol, U75354) . Mononwrium sp.. roth - 
steini gp. (ABZI.07. \MSi'il). RhvtidoDonero 
victoriae (ACAS.Ol. U75350). and Vespula 
mgans' (ACAK.Ol. U75353). Vouchers are" 
placed in the Australian National Insect Col- 
lection. Canberra 

Naturwissenschaften 84 (1997) © Springer- Verlag 1997 

Table 1. Application of a relative rate test [10] to the sequence date discussed in Fig. 1 

K|2-K| 3 



































0.53645 1 




















0.03068 1 








The Vespitlii sequence was used as the outgroup. The columns show the difference between the 
distances to the outgroup. the standard deviation of the difference between the two distances, 
and the Z-stalistic. None of the values approach the significance level (Z= 1.96) and hence the 
BonfeiToni correction |17] was not necessarv. .A larger test including an additional 14 Myrme- 
ciii sequences |5| also showed no detectable departure from a molecular clock. 

era of the first hymenopteran fossils 
[14|. Such a placement suggests that 
the ants diversified more slowly than 
previously thought, and that early di- 
versification of the order Hymenop- 
tera may have been more rapid than 
previously believed. The ants thus ar- 
ose very early during the breakup of 
Pangaea [15. 16]. indicating that ma- 
jor features of current distributions 
may reflect the order of separation of 
the continents. 

We thank P. Cranston. A. Gladow. 
A. A. Hoffmann and P. Patnilo for 

helpful suggestions in the course of 
this work. S. Darby for Vespida speci- 
mens. S.O. Shattuck for ant identifi- 
cations, the Australian Research 
Council and the Ian Potter Foundation 
for research support to R.H.C.. and 
the John Curtin School of Medical 
Research at the Australian National 
University for its support of L.S.J. 

1. Wilson. E.O.: .Social insects as dominant 
organism, in: Biology and Evolution of 
.Social Insects. J. Billen ed.. Leuven: Bel- 
aium. Leuven Universitv Press 1992. 1-7 








Holldobler. B.. Wilson. E.O.: The Ants. 
Cambridge. MA: Harvard University 
Press 1990 

Brandao. C.R.F.. Martins-Neto. R.G.. 
Vulcano. M.A.: Psyche 96, 195-208 

Crosland. M.W.J.. Crozier. R.H.: 
Science 2i/. 1278 (1986) 
Crozier. R.H.. Dobric. N.. Imai. H.T.. 
Graur, D.. Comuet. J.-M.. Taylor, R.W.: 
Mol. Phylogenet. Evol. 4. 20-30 (1995) 
Saccone. C, Lanave. C Pesole. G.. 
Preparata. G.: Methods Enzymof 183, 
570-583 (1990) 

Kimura. M.: J. Mol. Evol. 16. 111-120 

Harland. W. B.. Armstrong. R. L.. Cox, 
A. v., Craig. L.E.. Smith. A.G.. Smith. 
D.G.: A Geological Time Scale 1989. 
Cambridge. UK: Cambridge University 
Press 1990 

Wu. C.-l.. Li. W.-H.: Proc. Natl. Acad. 
Sci. USA S2. 1741-1745 (1985) 
Muse. S.V.. Weir. B.S.: Genetics 132. 
269-276 (1992) 

Jermiin. L.S.. Graur. D.. Lowe. R.M.. 
Crozier. R.H.: J. Mol. Evol. 39. 160- 
173 (1994) 

Crozier. R.H.. Crozier. Y.C.: Genetics 
133. 97-117 (1993) 

Pamilo. R. Bianchi. N.O.: Mol. Biol. 
Evol. 10. 271-281 (1993) 
Carpenter. F. M.. Bumham. L.: Annu. 
Rev. Earth Planet. Sci. 13. 297-314 

Windley. B.F.: The Evolving Conti- 
nents. Chichester: Wiley 1986 
Briggs. J.C.: Biogeography and Plate 
Tectonics. Amsterdam: Elsevier 1987 
Rice. W.R.: Evolution 43. 223-225 

Jermiin. L.S.. Crozier. R.H.: J. Mol. 
Evol. .?«. 282-294 (1994) ^ 

Naturwissenschatfen 84. 23-25 (1997) £) Springer- Veriag 1997 

Heterogeneous Turnover of Molecular Organic 
Substances from Crop Soils as Revealed by ^^C 
Labeling at Natural Abundance with Zea mays 

Eric Lichtfouse * 

Laboratoire de Biogeochimie Isotopique. Universite Pierre et Marie Curie. 
F-75252 Paris 

Because of the recent suggestion that 
the terrestrial biosphere may contain a 

* Present address: Laboratoire Sols et Envir- 
onnement. INRA/ENSAIA-INPL. BP 172. F- 
54505 Vandoeuvre-les-Nancv. France 

missing sink of atmospheric CO2 [IJ. 
studies on the dynamics of soil organ- 
ic carbon are of particular interest to 
try to understand changes in the glob- 
al carbon cycle [2]. Most investiga- 
tions build models that consider soil 

carbon as the sum of bulk pools of 
various turnovers [2]. However, a 
simple consideration of the huge reac- 
tivity difi'erences that exist between 
individual organic substances, such as 
lipids, amino acids, and carbohy- 
drates, strongly suggests that these 
compounds may also have signifi- 
cantly different dynamics in soils. If 
this is the case, models should take 
into account such heterogeneities at 
the molecular level. However, the 
dearth of knowledge on the dynamics 
of molecular species stems in large 
part from the extreme complexity of 
the soil medium and from the lack of 
suitable methods that allow long-term 
kinetics of organic species to be eval- 

Naturwissenschaften 84 (1997) © Springer- Veriag 1997